CAPITULO 5 RESULTADOS Y DISCUSIÓN
5.2 Espectroscopia de Rayos X
5.2.1 Caracterización por Difracción de Rayos X
the seasonal behaviour of a subsocial insect?
110
Abstract –
Cuticular hydrocarbons (HC) have been increasingly observed to provide a complex source of information that mediate numerous behaviours between individuals including species and kin recognition, sex determination, social dominance and reproductive status. The European earwig, Forficula auricularia L. (Dermaptera: Forficulidae) is a cosmopolitan insect species found in many temperate regions worldwide. Earwigs exhibit a complex life-cycle, which includes maternal care, aggregation behaviours and the formation of mating pairs in subterranean nests prior to over-wintering. Seasonal earwig population monitoring has also demonstrated that earwig trap catches rapidly decline after their final juvenile moult. Recently, unsaturated cuticular HCs have been implicated in earwigaggregation behaviours. We investigate whether this decline in earwig trap catches is linked to fluctuations in earwig cuticular HC profiles and whether this decline relates to the differing behaviours in the field. This was achieved by sequentially sampling field collected earwigs over a 21 week field season and quantifying 51 cuticular HCs using gas-chromatography mass-spectrometry, while monitoring the seasonal decline in earwig trap catches. Our results show that earwig cuticular HCs do indeed fluctuate throughout their activity season. In female earwigs, the concentration of long-chain methyl-branched HCs greater than 27 carbon atoms in length increased > 1000-fold toward over-wintering. In males, these compounds were observed to diminish. We also demonstrate that production of the unsaturated cuticular HCs, (Z)-9-tricosene, (Z)-7-tricosene, (Z)-9-pentacosene,(Z)-7-pentacosene and (Z,Z)-6,9- pentacosadiene, which have previously been hypothesised to be F. auricularia‟s aggregation pheromone components declined in both sexes from (mean ± SEM) 137.6 ng (± 30.9) in newly moulted males to 3.1 ng (± 0.8) in over-wintering individuals and from 37.3 ng (± 3.8) in newly moulted females to 1.4 ng (± 0.5) in over-wintering females. We also demonstrate that this decline in unsaturated HC production correlates strongly with the decline in earwig trap catches. We discuss whether the decline in earwig population estimates may be
potentially linked to the timing of the formation of mating pairs and subsequent subterranean nesting behaviours, which may begin earlier in the season than previously reported.
Key Words - Forficula auricularia, Dermaptera, Aggregation pheromone, Plasticity, Aging, Alkenes, Methyl-branched hydrocarbons
111
INTRODUCTION
The European earwig, Forficula auricularia L. (Dermaptera: Forficulidae) is an invasive insect pest, native to Europe, western Asia and possibly Northern Africa (Lamb and Wellington 1975). Several accidental introductions in both the northern and southern hemispheres have led to its successful establishment in many temperate regions world-wide (Lea 1903; Crumb et al. 1941; Guillet et al. 2000b). Previously, climate and locality were believed to affect F. auricularia life-history (Wirth et al. 1998). High altitude populations were observed laying one clutch per season during early winter with a long gregarious adult phase and no diapause and those at lower altitudes laying two clutches per season with an imaginal overwintering diapause (Guillet et al. 2000a), the first clutch being laid at the beginning or end of winter with a second smaller clutch in late spring early summer (Lamb and Wellington 1975). However, genetic analysis of populations in Europe and North America identified two subspecies; subspecies A (laying one or two clutches per year) and subspecies B (laying two clutches per year) (Guillet et al. 2000a). Studies have also demonstrated that these populations co-exist in the wild with forced copulations between subspecies in the laboratory showing egg infertility that prohibits any genetic flow occurring between populations (Wirth et al. 1998).
In addition to these differences in reproductive strategy, F. auricularia displays various complex behaviours within its lifecycle, which differentiate it from other insect taxa
including the formation of mixed aggregations containing both adult sexes and juveniles and maternal care (Lamb and Wellington 1975; Lamb 1976; Walker et al. 1993; Helsen et al. 1998). These behaviours have led to earwigs being increasingly considered a prime insect model to study the evolution of insect behaviours (Tomkins and Simmons 1998; Tomkins and Brown 2004; Mas and Kölliker 2011a; Mas and Kölliker 2011b).
In late autumn, male and female earwigs form pairs and excavate subterranean nests > 2 cm beneath the soil surface or under rocks and logs in preparation for overwintering (Lamb and Wellington 1975). Mating begins from late summer (Lamb and Wellington 1975) and
continues through the overwintering phase (S. Quarrell, pers. obs.). Multiple mating has been observed in laboratory experiments but it remains unclear whether this occurs in field
populations (Lamb 1976; Walker and Fell 2001). As mating may occur prior to nesting the male may not be the contributor of the paternal line (Brown 2006). Following nest formation
112 and mating the male exhibits mate guarding behaviours to prevent sneaky matings from other males and ensure paternity (Lamb 1976; Brown 2006). High male mortality is commonly observed during this over-wintering phase (Lamb 1976; Gingras and Tourneur 2001). Egg laying occurs mid to late winter, with any surviving males then aggressively evicted from the nest by the females soon after oviposition, after which time these males soon die (Lamb and Wellington 1975; Lamb 1976).
Female earwigs show strong maternal care for both eggs and young nymphs with eggs turned and cleaned to limit fungal infection (Kolliker and Vancassel 2007). Brooding females provide food throughout the first nymphal instar via two behavioural mechanisms either food regurgitation or by direct provisioning i.e. whole aphids (Staerkle and Kolliker 2008). First instar nymphs remain in the nest with the female until the end of the first moult, when both nymphs and females leave the nest to either nocturnally feed on vegetation and other insects then returning to the nest by day or leaving the nest permanently (Lamb and Wellington 1975). At this point in time the females of subspecies A will die and the females of subspecies B will establish another nest and lay again (Lamb and Wellington 1975). In orchards and forested areas free foraging earwigs are predominantly arboreal with earwigs residing under rocks, logs and within leaf litter where trees are not present (Lamb and Wellington 1975).
During this free foraging phase, earwigs form mixed aggregations that contain both adult sexes and all life stages, which are mediated via the use of an aggregation pheromone (Sauphanor 1992; Walker et al. 1993; Hehar 2007). However, these studies have failed to isolate the pheromone. One notable omission from these aggregation pheromone studies are the numerous cuticular hydrocarbons (HC) identified from female earwig cuticles by Liu (1991). Recently cuticular HCs were shown to be involved with maternal care behaviour, in particular, food provisioning to juveniles with juvenile HC composition fluctuating when food quality/ quantity was altered. These fluctuations were demonstrated to impact on the maternal care behaviour of nesting females (Mas et al. 2009; Mas and Kölliker 2011a) and the timing of future reproductive events (Mas and Kölliker 2011b). More recently, the cuticular HCs (Z)-9-tricosene, (Z)-7-tricosene, (Z)-9-pentacosene,(Z)-7-pentacosene have been implicated as the compounds, which mediate F. auricularia aggregations (see Chapter 5).
113 After the final juvenile moult, a rapid decline in adult earwig numbers in monitoring traps has been observed (Quarrell 2008; Moerkens et al. 2009). Moerkens et al. (2009) postulated that this decline reflects a real drop in the earwig population mediated by density dependent factors including reduced food availability, increased natural enemy populations, disease or the use of insecticides. However, this study was unable to confirm any of these hypotheses. An alternate cause for this decline, which was not hypothesised, is that high numbers of earwigs in traps is promoted by the active production of the aggregation pheromone and that the trap declines reflect the switching off of this pheromone interlinked with the formation of mating pairs earlier in the season than previously thought in the field. If this is the case this process may well be endocrine regulated as hormones have been shown to control insect reproductive cycles, species migration and pheromone production in insects (Barth 1965; Dukas and Mooers 2003; Schal et al. 2003). Indeed, juvenile hormone (JH) has already been shown to regulate the sexual maturity, reproductive cycles and maternal care instincts in the earwigs; Euborellia annulipes Lucas (Rankin et al. 1995a; Rankin et al. 1995b; Rankin et al. 1997), Labidura riparia Pallas (Baehr et al. 1982; Vancassel et al. 1984) and Anisolabis maritima Bonelli (Rankin et al. (1995a) cites Ozaki (1960)) and therefore may also regulate the production of F. auricularia‟s aggregation pheromone. If the aggregation pheromone used by earwigs does decline throughout the season it may explain both the seasonal decline in earwig trap catches and the difficulty observed in isolating the earwig aggregation
pheromone (Walker et al. 1993; Hehar 2007).
The aims of this study were to identify if any temporal fluctuations in the cuticular HCs of F. auricularia do occur throughout the earwig activity season, and to determine if these
fluctuations correlate with the observed decline in earwig trap catches in the field and any changes in earwig behaviour. To do so we sequentially sampled the cuticular HCs of field- based earwigs while simultaneously monitoring earwig population dynamics.
METHODS AND MATERIALS
Chemical analysis
Six male and six female earwigs were collected from apple trees within a commercial apple orchard in the Huon Valley, Tasmania (Lat. 42˚ 59.755' S Long. 147˚ 4.328' E) every two weeks between the 16th December 2011 and the 20th March 2012. The earwigs were
114 collected by placing twenty corrugated cardboard rolls (8.5 cm x 9 cm) attached with garden twine (Zenith, REA 0060), at the base of each tree 30 cm above ground level. All earwigs were randomly selected from a variety of traps at each time point. Over-wintering individuals were collected from subterranean nests at the same site on the 9th May 2012.
Cuticular HCs were identified and quantified by immersing whole earwigs in 1 ml of hexane containing an n-C22 HC standard (50 µL; 25 µg in 1 ml) for one hour. Solvent extractions were then reduced under a gentle flow of nitrogen to ca. 100 µL and transferred into 150 µL Waters inserts (WAT 094171) for GC-MS analysis and stored at -6 °C until required. GC-MS analysis of hexane washes was performed with a Varian CP 3800 gas chromatograph, fitted with a Varian VF5-MS column (30 m, 0.25 mm, 0.25 um film thickness) coupled to a Bruker 300-MS triple quadrupole mass spectrometer in electron ionisation mode using 70 eV electrons. Samples were injected with a Varian CP-8400 autosampler into a Varian 1177 split/splitless injector at 270 °C with a 30:1 split ratio. Oven temperature was programmed from 50 ˚C (2 minute hold) to 150 ˚C at 30 ˚C per minute, then 150 ˚C to 300 ˚C at 8 ˚C/min (1 minute hold). Carrier gas flow was helium at 1.2 ml/minute using a constant flow mode. The MS was scanned from m/z 35 to 600 at 3 scans per second.
Methyl-branched hydrocarbons were identified as per chapter 5 using n-alkane standards, mass spectral fragmentation patterns from Doolittle et al. (1995) and Kroiss et al. (2011) and published retention index data from Carlson et al. (1998) and Katritzky et al. (2000). Double bond positions from alkenes and alkadienes were identified by derivatisation with dimethyl disulfide (DMDS) as per Carlson et al. (1989). Alkatriene double bond positions were
determined using underivatised samples via mass spectral fragmentation patterns as described by Miller (2000) and Conner et al. (1980).
Earwig population monitoring
Earwig populations were monitored in a neighbouring apple block (ca. 50 m away) using corrugated cardboard rolls (8.5 cm x 9 cm) and at the same time points as stated above. The number, sex and life stage of each earwig found in the cardboard rolls was recorded and subsequently released at the tree base. The earwig traps were replaced fortnightly to prevent the aggregation pheromone from permeating into the trap and falsely inflating earwig population monitoring efforts (see chapter 5).
115
Statistical Analysis
Freidman‟s tests with Dunn-Bonferroni multiple pair-wise comparisons tests were performed on the earwig population data. To assess changes in individual cuticular HCs throughout the observation period Kruskal-Wallis tests with Dunn-Bonferroni multiple pair-wise
comparisons tests were performed. Both Freidman‟s and Kruskal-Wallis tests were conducted using IBM SPSS Statistics version 19. To determine whether a correlation exists between the production of the HCs quantified and those behaviourally tested in Chapter 5 and the earwig trapping data Spearman‟s rho was also performed. To establish whether a relationship exists between the production of any single cuticular HC within the complete profile throughout the observation period and the number of earwigs found aggregating in trees, recursive
partitioning was also performed by analysing the 51 HCs quantified together with the earwig trap catch data. Recursive portioning develops conditional inference trees. At each step a null hypothesis of no association is tested between the outcome and the covariates with the
processing stopping if the null hypothesis is retained. If the null hypothesis is not retained the covariate with the strongest association is used to split the data into disjoint sets. This process is repeated until no covariate is associated with the data set. Recursive partitioning was performed using R version 2.15.1 using the “party” package and the “ctree” function.
RESULTS
Earwig population dynamics
The number of male, female and all juvenile earwigs observed in traps varied significantly over the field season (Figure 6-1A and 6-1B; Friedman‟s male χ2
= 60.60, P < 0.001; female χ2
= 57.32, P < 0.001; 4th instars χ2 = 133.15, P < 0.001; 3rd instars χ2 = 69.80, P < 0.001 and 2nd instars χ2 = 43.78, P < 0.001). The earwig population displayed characteristics of
subspecies B (Wirth et al. 1998) where two generations of 4th instar juveniles are apparent peaking in numbers at or prior to the commencement of the observation period at week 1 with a second smaller generation peaking at ca. week 7 (Figure 6-1A). Male and females both peaked in number on the week 5 where a mean (± SEM) of 2.05 ± 0.46 males (range 0 – 9) and 2.35 ± 0.53 female earwigs (range 0 – 9) were observed per tree. A second smaller peak in adult numbers, though not statistically significant was also observed after the apple harvest in week 17 (Figure 6-1B; Bonferroni adjusted: males Z = 3.003, P = 0.120; females Z = 2.977, P = 0.131), which then significantly diminished by week 19 (Bonferroni adjusted: males Z = 3.525, P = 0.019; females Z = 3.865, P = 0.005).
116
Figure 6-1. Mean (± SEM) Forficulaauricularia per trap collected from apple trees (n = 20) from the 16th December 2011 to 5th May 2012 A) 2nd, 3rd and 4th instars earwigs per trap B) Adult male and female earwigs per trap.
HC analysis
A total of 51 cuticular HCs were identified from the hexane washed cuticles of male earwigs comprising alkanes, alkenes, alkadienes and alkatrienes varying from 21 to 31 carbon atoms in length and 49 HCs from the cuticles of female earwigs with neither of the alkatrienes, (Z,Z,Z)-3,6,9-C25 or(Z,Z,Z)-3,6,9-C27 recorded from the female cuticles (Table 6-1, Figures 6-2 and 6-3). The total HC concentration of adult male and female earwigs declined
significantly between the start of the monitoring period in both males (Figures 6-2 and 6-3; Kruskal-Wallis χ2 = 35.45, df = 9, P < 0.001) and females (Kruskal-Wallis χ2
= 35.05, df = 9, P < 0.001). In males, the concentrations of only one HC, 3,7-diMe-C25 was observed not to have changed throughout the field season (Figure 6-3, Appendix 2; Kruskal Wallis; males χ2
0 5 10 15 20 25 30 4th instar 3rd instar 2nd instar 0 1 2 3 4 5 1 3 5 7 9 11 13 15 17 19 21
Collection date (weeks)
Male Female
A
B
M
ea
n e
a
rw
ig
s/
trun
k trap
117 = 12.906, df = 9, P = 0.167). In females the production of this compound was observed to fluctuate significantly (χ2
= 24.270, df = 9, P = 0.004) where elevated levels were observed in early season (mean ± SEM; week 1 0.031 µg ± 0.008, week 3; 0.112 ± 0.077), late season (week 19; 0.156 ± 0.075) and in subterranean females (week 21; 0.125 ± 0.053). Despite the total HC profile concentration in females declining numerous long-chain methyl-branched HCs greater than 27 carbon atoms in length were observed to increase over 1000 times in concentration (Figure 6-3, Appendix 2)
Cuticular washes from recently moulted adult males collected in week 1 possessed significantly more HC (Figure 6-2; Mann-Whitney; Z = -2.611, P = 0.008) than recently moulted females (mean, n-C22 equivalents ± SEM; males 192.5 µg ± 3.4; females 77.2 µg ± 2.4). However, this trend was not consistently observed throughout the season with over- wintering males collected from within subterranean nests on week 21 having less HC (5.3 µg ± 0.7) than over-wintering females which contained 36.9 µg (± 9.8) of HC (Figure 6-2; Mann-Whitney; Z = -2.562, P = 0.009). Similarly, in week 15 males possessed significantly less HC than females (Mann-Whitney; Z = -2.562, P = 0.009) where males possessed (mean ± SEM) 7.971 µg ± 0.190 of HC compared to females which possessed 10.963 µg ± 0.251. Conversely, in week 17 a two-fold decline in HC was recorded in females resulting in males possessing more HC than females (Mann-Whitney; Z = -2.402, P = 0.015) with males possessing (mean ± SEM) 7.188 µg ± 0.193 of HC compared to females which possessed 5.033 µg ± 0.115. There was no significant differences in the quantity of cuticular HC between sexes in any of the other weeks (Mann-Whitney; week 3 Z = -0.183, P = 0.931; week 5 Z = -1.826, P = 0.082; week 7 Z = -1.441, P = 0.180; week 9 Z = -1.121, P = 0.310; week 11 Z = -0.961, P = 0.394; week 13 Z = -0.183, P = 0.931; week 19 Z = -0.873, P = 0.937).
The large quantities of two alkatrienes, (Z,Z,Z)-3,6,9-C25 (peak 13) and (Z,Z,Z)-3,6,9-C27 (peak 26) produced by males, but not females, was observed in its highest concentrations in week 1 (Figure 6-2; 0.17 µg ± 0.06 and 0.54 µg ± 0.07 respectively). However, the
production of these compounds declined significantly throughout the season (Kruskal-Wallis: (Z,Z,Z)-3,6,9-C25 χ2 = 30.492, df = 9, P < 0.001; (Z,Z,Z)-3,6,9-C27 χ2 = 37.596, df = 9, P < 0.001). The majority of the decline in these compounds were observed in the four weeks after the final moult has occurred (Appendix 2; Bonferroni adjusted, (Z,Z,Z)-3,6,9-C25 Z = 4.129, P = 0.002; (Z,Z,Z)-3,6,9-C27 Z = 4.684, P < 0.001) with subterranean males at the end of the
118 season possessing 0.003 µg (± 0.001) and 0.006 µg (± 0.002) for both (Z,Z,Z)-3,6,9-C25 and (Z,Z,Z)-3,6,9-C27 respectively. The unsaturated HCs (Z)-7-C25 and (Z)-7-C27 were both observed to increase, though not significantly within the first 2 weeks after the final moult has occurred (Figure 6-4, Bonferroni adjusted; males (Z)-7-C25 Z = -1.850, P = 1.00; (Z)-7- C27 Z = -0.995, P = 1.00 ; females (Z)-7-C25 Z = 0.304, P = 1.00; (Z)-7-C27 Z = -0.458, P = 1.00). For all HC phenology data see Appendix 2.
119
Table 6-1. Complete list of compounds detected from the cuticles of F. auricularia.
Peak Compound name Abbreviation Peak Compound name Abbreviatio
n
1 n-heneicosane n-C21 27 (Z)-7-heptacosene (Z)-7-C27
2 (Z)-9-tricosene (Z)-9-C23 28 n-heptacosane n-C27
3 (Z)-7-tricosene (Z)-7-C23 29 15-methylheptacosane 15Me-C27
4 n-tricosane n-C23 29 13-methylheptacosane 13Me-C27
5 7-methyltricosane 7Me-C23 30 11-methylheptacosane 11Me-C27
6 5-methyltricosane 5Me-C23 31 9-methylheptacosane 9Me-C27
7 3-methyltricosane 3Me-C23 32 7-methylheptacosane 7Me-C27
8 n-tetracosane n-C24 33 5-methylheptacosane 5Me-C27
9 3,7-dimethyltricosane 3,7-diMe-C23 34 11,15-dimethylheptacosane 11,15-
diMeC27
10 6,9-pentacosadiene 6,9-C25 35 9,13-dimethylheptacosane 9,13-diMeC27
11 (Z,Z)-6,9-pentacosadiene (Z,Z)-6,9-C25 36 3-methylheptacosane 3Me-C27
12 (Z)-9-pentacosene (Z)-9-C25 37 n-nonacosane n-C29
13 (Z,Z,Z)-3,6,9-pentacosatriene (Z,Z,Z)-3,6,9-C25 38 15-methylnonacosane 15Me-C29
14 (Z)-7-pentacosene (Z)-7-C25 39 13-methylnonacosane 13Me-C29
15 n-pentacosane n-C25 40 11-methylnonacosane 11Me-C29
16 13-methylpentacosane 13Me-C25 41 9-methylnonacosane 9Me-C29
17 11-methylpentacosane 11Me-C25 42 7-methylnonacosane 7Me-C29
18 9-methylpentacosane 9Me-C25 43 11,15-dimethylnonacosane 11,15-
diMeC29
19 7-methylpentacosane 7Me-C25 44 9,13-dimethylnonacosane 9,13-diMeC29
20 5-methylpentacosane 5Me-C25 45 3-methylnonacosane 3Me-C29
21 3-methylpentacosane 3Me-C25 46 15-methylhentriacontane 15Me-C31
22 n-hexacosane n- C26 47 13-methylhentriacontane 13Me-C31
23 3,7-dimethylpentacosane 3,7-diMe-C25 48 11-methylhentriacontane 11Me-C31
24 6,9-heptacosadiene 6,9-C27 49 9-methylhentriacontane 9Me-C31
25 (Z)-9-heptacosene (Z)-9-C27 50 11,15-dimethylhentriacontane 11,15-
diMeC31
120
Figure 6-2. Representative gas-chromatograms of Forficula auricularia cuticular hydrocarbons collected from A) a recently moulted male B) an over- wintering male collected from a subterranean nest C) a recently moulted female D) an over-wintering female collected from a subterranean nest. Numbers above peaks refer to compounds listed in Table 6-1.
A
B
C
D
121
Figure 6-3. Mean percentage change in Forficula auricularia cuticular HC composition between recently moulted and over-wintering adult A) males and B)
females. Six male and six female earwigs were collected and analysed at each time point. Negative values indicate a decline in HC production. Positive values indicate an increase in production. All HCs were observed to change over time unless otherwise indicated (Kruskal-Wallis; P < 0.05). NS indicates no
significant difference. For all HC quantities (µg) and P-values see Appendix 2.
-150 -100 -50 0 50 100 -500 0 500 1000 1500 2000 2500 3000 3500 n -C 21 (Z) -9- C2 3 (Z) -7- C2 3 n -C 23 7M e- C 23 5M e- C 23 3M e- C 23 n -C 24 3,7d iMe -C 23 6,9- C2 5 (Z, Z)- 6,9- C2 5 (Z) -9- C2 5 (Z, Z, Z)- 3,6, 9- C2 5 (Z) -7- C2 5 n -C 25 13 Me -C2 5 11 Me -C2 5 9M e- C 25 7M e- C 25 5M e- C 25 3M e- C 25 n -C 26 3,7d iMe -C 25 (Z, Z)- 6,9- C2 7 (Z) -9- C2 7 (Z, Z, Z)- 3,6,